Statistical quality assessment of fingerprints

• US 7,489,807 B2
• Filed: 08/06/2004
• Issued: 02/10/2009
• Est. Priority Date: 08/07/2003
• Status: Expired due to Fees

• Current Assignee: L-1 Secure Credentialing, LLC (Advent International Corporation)
• Original Assignee: L-1 Secure Credentialing, LLC (Advent International Corporation)
• Inventors: Hwang, Kyungtae, Kim, Nam-Deuk
• Primary Examiner(s): Mariam; Daniel G
• Application Number:
  US10/913,015
• Publication Number:
  US 20050069179A1
• Time in Patent Office: 1,649 Days
• Field of Search: 382/115, 382124-125, 382/218, 382/294, 283 68- 70
• US Class Current: 382/124
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: G06V 40/1359 Extracting features related to ridge properties; Determining the fingerprint type, e.g. whorl or loop
• CPC Class Codes: G06V 40/1371 Matching features related to minutiae or pores
• CPC Class Codes: G06V 40/53 Measures to keep reference information secret, e.g. cancellable biometrics

Abstract

The present invention relates generally to human fingerprints. More specifically the present invention relates to assessing the quality of a fingerprint image. An assessment is made by analyzing characteristics of image sub-regions. For example, the characteristics may include statistics that are associated with a sub-region. If a sub-region is found to have unacceptable characteristics, it can be modified through a morphological operation to compensate for the unacceptable characteristics.

Claims

1. A method to analyze a fingerprint image corresponding to a human fingerprint of a human subject, the fingerprint image including ridges and valleys, said method comprising:

receiving a fingerprint image corresponding to the human fingerprint of the human subject;

determining image distortion associated with the fingerprint image;

adjusting the fingerprint image to compensate for the image distortion;

segmenting the fingerprint image into at least a plurality of subsets; and

for each of the subsets;

determining a measure associated with fingerprint ridges and fingerprint valleys in the subset; and

based on the measure assigning the subset to at least one of a plurality of categories;

assessing an overall acceptability of the fingerprint image by analyzing at least one of the plurality of categories, and requiring capture of another fingerprint image from the human subject if the overall acceptability of the fingerprint image is at or below predetermined criteria.

2. The method of claim 1 wherein said analyzing comprises determining whether a predetermined number of subsets are assigned to a first of the plurality of categories.

3. The method of claim 2, wherein the first category comprises at least one of acceptable, wet or dry.

4. The method of claim 2, further comprising accepting the fingerprint image when a predetermined number of subsets are assigned to the first of the plurality of categories.

5. The method of claim 2 further comprising rejecting the fingerprint image when a predetermined number of subsets are not assigned to the first of the plurality of categories.

6. The method of claim 1, wherein the measure comprises a statistical measure.

7. The method of claim 1, wherein said determining image distortion associated with the image data comprises determining a global orientation.

8. The method of claim 7, wherein said determining image distortion associated with the image data further comprises determining an orientation relative to an image subset.

9. The method of claim 1, wherein the plurality of subsets comprises less image data than the received fingerprint image.

10. The method of claim 1, further comprising filtering the image to remove noise.

11. The method of claim 1, further comprising modifying image data defined by a subset when the subset is assigned to a predetermined category.

12. The method of claim 11, wherein the modifying comprises adjusting ridge or valley dimensions.

13. A method of generating an identification document including a fingerprint image comprising:

receiving a fingerprint image that has been analyzed according to claim 1; and

printing the fingerprint image on the identification document or storing the fingerprint image in memory provided by the identification document.

14. The method of claim 13, wherein the memory comprises at least one of electronic memory circuits, optical memory or magnetic memory.

15. A method to analyze a fingerprint image, the fingerprint image including ridges and valleys, said method comprising:

receiving a fingerprint image corresponding to a human fingerprint;

determining image distortion associated with the fingerprint image;

adjusting the fingerprint image to compensate for the image distortion;

segmenting the fingerprint image into at least a plurality of subsets;

for each of the subsets;

determining a measure associated with fingerprint ridges and fingerprint valleys in the subset; and

based on the measure assigning the subset to at least one of a plurality of categories;

determining which of the subsets correspond to a fingerprint core, and weighting those subsets that correspond to the fingerprint core more significantly relative to those subsets that do not correspond to the fingerprint core; and

assessing an overall acceptability of the fingerprint image by analyzing at least one of the plurality of categories.

16. A method to analyze a fingerprint image, the fingerprint image including ridges and valleys, said method comprising:

receiving a fingerprint image corresponding to a human fingerprint;

determining image distortion associated with the fingerprint image;

adjusting the fingerprint image to compensate for the image distortion;

segmenting the fingerprint image into at least a plurality of subsets;

for each of the subsets;

determining a statistical measure associated with fingerprint ridges and fingerprint valleys in the subset, wherein the statistical measure comprises at least a standard deviation of ridge thickness and a standard deviation of valley thickness; and

based on the statistical measure assigning the subset to at least one of a plurality of categories.

17. The method of claim 16, wherein the statistical measure further comprises at least one of mean of ridge thickness or mean of valley thickness.

18. A method of determining whether to accept a first fingerprint image or to require capture of a second fingerprint image, the first fingerprint and second fingerprint images comprising a representation of a same human fingerprint, the fingerprint including ridges and valleys, said method comprising:

determining an orientation of the first fingerprint image;

adjusting the fingerprint image according to the orientation;

determining statistical values that are associated with at least some of the ridges and valleys;

comparing the determined statistical values against predetermined acceptance criteria;

and determining whether to accept the first fingerprint image based at least on the comparison.

19. The method of claim 18, wherein statistics are determined for selected image areas.

20. The method of claim 18, wherein the statistical values are determined based on ridge and valley thickness.

21. The method of claim 18, wherein the statistical values are determined based on a relationship between the ridges and valleys.

22. The method of claim 18, wherein statistical values associated with a fingerprint core are considered to be more significant than statistical values that are not associated with the fingerprint core.

23. The method of claim 18, wherein the statistical values comprise a measure related to a frequency domain analysis of at least some of the ridges and valleys.

24. A method of generating an identification document including a fingerprint image comprising:

receiving a fingerprint image that has been processed according to claim 18; and

printing the fingerprint image on the identification document or storing the fingerprint image in memory provided by the identification document.

25. A method to analyze image data corresponding to a human fingerprint, the human fingerprint including fingerprint ridges and valleys, said method comprising:

receiving image data corresponding to a human fingerprint;

determining image distortion associated with the image data;

adjusting the fingerprint image to compensate for the image distortion;

segmenting the fingerprint image into at least a plurality of subsets; and

for each of the subsets;

determining a measure associated with the human fingerprint in the subset; and

based on the measure assigning the subset to at least one of a plurality of categories;

determining acceptability of the human fingerprint at least through use of or reference to at least one of the plurality of categories, and requiring recapture of the human fingerprint if the acceptability of the human fingerprint is at or below predetermined criteria.

26. A computer readable medium comprising instructions or circuitry to perform the method of claim 25.

27. A method of determining whether to accept a first fingerprint image or to require capture of a second fingerprint image, the first fingerprint and second fingerprint images comprising a same representation of a human fingerprint, the fingerprint including ridges and valleys, said method comprising:

determining an orientation of the first fingerprint image;

adjusting the fingerprint image according to the orientation;

determining statistical values that are associated with at least a portion of the human fingerprint;

comparing the determined statistical values against predetermined acceptance criteria;

and determining whether to accept the first fingerprint image based at least in part on the comparison.